Dynamic Cone-Beam Reconstruction Using a Variational Level Set Formulation

Tomographic reconstruction from cone-beam X-ray data is only solved for static objects, e. g. the abdomen. In cardiac imaging, a rotational angiography sequence takes approx. 5 s and therefore spans several heart beats. Since such an acquisition scheme forces a trade-off between consistency of the scene and reasonable angular spacings between camera positions, standard reconstruction techniques fail at recovering the 3D + t scene. We propose a new reconstruction framework based on variational level sets including a new energy term for symbolic reconstruction and incorporating the motion into the level set formalism. The resulting simultaneous estimation of shape and motion proves feasible in the presented experiments. Since the proposed formulation offers a great flexibility in incorporating other data terms, it could be of interest for other reconstruction settings as well.